In multiple sclerosis (MS), entry of inflammatory leukocytes and humoral factors into the CNS drives the onset of clinical disease. Permanent damage can be reduced by limiting ingress of pathogenic lymphocytes and monocytes, and soluble factors including antibodies and serum proteins. This project seeks to identify CNS mechanisms controlling cell and humoral entry in MS lesions, to define new therapeutic strategies to limit damage and disability. To enter the CNS from the vasculature, cells and soluble factors must traverse two distinct barriers. Initially, they must cross the endothelial blood-brain barrier (BBB) into the perivascular space (PVS). Then, to enter the parenchyma they must penetrate the astrocytic glia limitans (GL). The first funding period of this award focused on the BBB. We identified reactive astrocytes as key drivers of BBB opening in lesions, acting via the permeability factors Vegf-a and Ecgf1. The BBB is sealed by endothelial tight junctions (TJ) of claudin (Cldn)-5 and occludin (Ocln) subunits. Vegf-a and Ecgf1 repress both, opening the BBB. Their blockade prevents BBB disruption, limiting pathology and disability. Here, we now focus on the second key barrier, the astrocytic GL, which controls CNS access upon BBB disruption. Critically, our data reveal that following astrocyte-driven opening of the BBB, these cells then control entry at the GL by forming TJ of their own. In inflammatory lesions, reactive astrocytes at the GL express Cldn-1/4, junction adhesion molecule-a (Jam-a), and Ocln-like tricellulin (Tcl) - the same TJ proteins which tightly seal skin and bladder epithelia. These results suggest astrocytes first drive BBB disruption in lesions, then act as gatekeepers to the CNS, at the GL. Importantly, pilot data further propose differential roles for astrocyte TJ proteins in control of cell vs. humoral entry. Claudins drive TJ strand formation, while Tcl may regulate strand complexity. Jam-a regulates function via intracellular signaling. Our results suggest that in astrocytic Cldn-4 knockout mice, leukocytes may not accumulate in the PVS in lesions, but directly transit the GL into the CNS. Conversely, humoral entry may be more severe in Jam-a conditional mice. Moreover, our data also propose leukocytes target astrocytic Cldn-4 for degradation, to facilitate transmigration. Here, we will test the hypothesis that reactive astrocytes control inflammatory lesion expansion via formation of a TJ barrier at the glia limitans. We propose three Aims. In Aim 1, we will test the extent to which astrocytic Cldn4 and Jam-a differentially regulate cellular vs. humoral trafficking. This work uses focal lesions in conditional knockout mice. In Aim 2, we will use culture models to identify the critical molecular mechanisms controlling leukocyte vs. humoral entry. In Aim 3, we will identify key leukocyte proteases targeting astrocytic Cldn-4 at the GL. We will then test the extent to which astrocytic TJ control disease severity in an MS model, via genetic targeting and therapeutic blockade. This proposal directly addresses control of leukocyte and humoral entry, and severity of neurologic deficit in MS. It also has broader implications for CNS access, drug delivery, and inflammatory disease more widely. The goal of this work is to identify novel therapeutic avenues to restrict lesion expansion and clinical exacerbation in patients.